A retrospective study was conducted to assess the appropriateness of the utilization of therapeutic drug monitoring (TDM) services with regards to antiepileptic drugs (AEDs) at a tertiary care hospital, and analysis of 5094 samples of the carbamazepine (CBZ), phenytoin (PHT), valproic acid (VAP), and phenobarbitone (PB) was undertaken. Maximum requisitions were received for CBZ (54.92%) followed by PHT (27.05%), VAP (14.40%), and PB (3.61%). About 2.12% requisitions were received for patients taking unlabeled AEDs. Reasons for TDM were routine monitoring (36.3%), adverse drug reactions (ADRs) (17.2%), relapse (30.7%), no response (3.35%), and irregular treatment (2.24%) and not mentioned (10.5%). Majority of the samples (69%) were drawn for measuring trough levels; however, in 22% time of last dose was not mentioned and 9.64% were for peak or random levels. In all, 6.9% requisitions for TDM were sent before steady-state levels, and in 15.5% duration of therapy was not mentioned. Blood levels within therapeutic range were found with CBZ (63%) followed by PB (56.52%), VAP (45.99%), and PHT (42.52%). Blood levels were above therapeutic range in 45.69%, 29%, and 21.73% patients taking VAP, PHT, and PB, respectively. Unsuspected poor compliance was uncovered in 11.8%, 41.2%, and 29.3% requisitions sent with ADR, relapse, and routine monitoring as reason for drug levels, respectively. Only half of all AED measurement requisitions were complete and met the criteria for appropriate AED-level determination. Incomplete requisitions lead to difficulty in uniform interpretation of results and thus add to unnecessary costs.